PRELIMINARY & UNOFFICIAL. Compiled from public internet sources (news, social media, USGS). Not a formal damage or risk assessment and not official figures. Verify before use. Method & sources
Situational overview · living page

M7.7 Azuero Earthquake, Panama

9 October 2026 · 12:56 local (17:56 UTC)
Live
Loading…

Key numbers

Official reported by Panamanian authorities   Reported compiled from media and social media   Estimate automated models, not counts

Impact map

Where the shaking was strongest and where impacts have been reported. Click a photo pin to see the damage, a summary and the source. Full-screen map →

How to read this map

Coloured bands show modelled shaking from the USGS ShakeMap (VI strong, VII very strong, VIII severe, IX violent). Photo pins are damage photos from news and social media, each checked by eye and placed at town level. Dots are other reported impacts. Circles are aftershocks, updated live from USGS (red = last hour). Switch on Potential building damage by district for the rapid exposure assessment. Locations are approximate. Few reports from an area can mean an information gap (power and telecom outages) rather than no damage.

At a glance

1 What happened

    2 What we know so far

      Preliminary, from news, social media and official statements. See all reports →

      3 What we don't know yet

        Reported impacts by area

        Services & infrastructure

        Latest damage photos

        All impact reports →

        Latest headlines (auto-refreshed)

          Full news stream →

          Behind the numbers

          Estimated deaths: probability by range

          USGS PAGER automated model from shaking and population. Not a count.

          Estimated economic losses (US$): probability by range

          USGS PAGER. Red alert: past events at this level needed a national or international response.

          Latest official statement

          Conflicting information

          Curated reports of impacts from news outlets, official statements and social media, each linked to its source. Confidence is our judgement of corroboration, not verification on the ground.

          Automated stream of news headlines (Google News, Spanish & English) and humanitarian reports mentioning the earthquake. Refreshed every 15 minutes; not curated — headlines may be inaccurate or duplicate.

          Rapid exposure assessment. How many people and buildings were exposed to each level of shaking: the USGS ShakeMap combined with population (WorldPop 2025, scaled to Census 2023) and building footprints (Microsoft ML footprints), by district. The potential damage figures apply generic damage rates per shaking level. They do not use building-type-specific vulnerability, so they are indicative only, not a loss model and not a substitute for field assessment (e.g. PDNA / GRADE).
          • Likely an underestimate in places. Public "Did You Feel It?" reports put shaking about one level higher than modelled near Penonomé (+1.5) and Santiago (+0.8), where damage is well documented. The "one level stronger" column shows that case.
          • Building coverage checked: Microsoft footprint counts are within ±9% of the GEM census-based building inventory in every affected province, so missing buildings are not the main source of error.

          People and buildings by shaking intensity

          Panama only. MMI class VII = 6.5–7.5.

          Potential buildings with moderate or worse damage (mid-point), by district.

          Districts ranked by potential building damage

          Damage ratios used

          Share of buildings reaching each damage state, by MMI class. Broad ranges for a mixed Central American stock (confined masonry and RC frames with a minority of adobe, quincha and unreinforced block). PAGER notes that the predominant vulnerable types in the region are unreinforced brick masonry and adobe.

          Secondary hazards

            Observed vs modelled — early signal

            The best-documented damage is in Penonomé (Coclé), where the ShakeMap estimates only MMI ~VI. That points either to stronger local shaking than modelled (site amplification, rupture directivity toward the north-east) or to the particular vulnerability of older heritage and commercial structures. Conversely, there are very few reports yet from southern Azuero (Tonosí, Macaracas, Las Minas, Los Pozos) where modelled shaking is strongest (MMI VIII–IX) — likely an information gap (power and telecom outages, rural areas) rather than an absence of damage.

            National risk context

            Building-vulnerability layer (census housing materials, GEM exposure model) in preparation.

            About this platform

            This platform was developed independently by Brenden Jongman to support rapid analysis of the October 2026 Panama earthquake and its potential impacts. It uses publicly available data and AI-assisted analytical and development tools.

            The information and estimates are preliminary, may contain errors, and should not be interpreted as validated damage or loss assessments. This is not an official World Bank product, and its contents do not necessarily reflect the views of the World Bank or its member governments.

            For preliminary analytical use only. Not for public distribution.

            Figures labelled "official" are those reported by Panamanian authorities as quoted in public sources; all other numbers are media reports or model estimates and will change.

            How it updates

            • Earthquakes and aftershocks are loaded live from the USGS API in your browser every 2 minutes.
            • News stream, USGS PAGER/ShakeMap status, GDACS are refreshed automatically every 15 minutes.
            • Curated impact reports, official figures and the overview text are reviewed and updated regularly by searching news, official channels and social media; every item links to its source.

            Data sources

              Known limitations

              • Media reports in the first hours are often incomplete, duplicated or wrong. Items are tagged by confidence; unconfirmed claims (e.g. trapped persons) are marked as such.
              • Locations are geocoded to town level and are approximate.
              • Areas with power and telecom outages are under-reported.
              • Exposure uses an automatic ShakeMap with a point-source rupture; it will change when USGS publishes a finite-fault model.
              • Building footprints from machine learning miss buildings under tree cover and in remote areas; counts are lower-to-mid estimates.
              • Images are hot-linked from their original publishers and remain their property.